Real ear sound pressure level : investigating a clinically feasible and individualised real ear correction for infants and adults undergoing audiological evaluation.

Abstract

Background: When hearing tests via electrophysiological assessment are performed on infants using insert earphones, smaller infant ear canal volumes relative to an adult mean that the stimulus generates a considerably higher sound pressure level than the dial setting on the test instrument. This effect is known as “infant uplift” and is brought about as audiometric scales (dB Hearing Level) are derived from and calibrated according to data from normally-hearing adults. To date, individualised corrections applied during infant hearing assessments do not account for the uplift. One approach to address this could be to measure the correction rapidly, using the microphone in a probe of an otoacoustic emissions (OAE) device. However, there is no existing data that compares the accuracy of such a swiftly-derived correction as compared with the gold-standard, but time-consuming, approach. To investigate the feasibility and application of the idea, the overarching objective of the present thesis is to perform such a comparison. Methods: A sample of 20 participants above the age of 18 years were recruited into the study. Data was collected monaurally, and exclusion criteria were contraindications for acoustical measurements. Using a counterbalanced designed, participants underwent a measurement of ear canal acoustics via the OAE probe check sequence and the probe microphone measurement. Procedures were performed twice to produce test-retest data. The level of agreement between the two techniques, and between test and re-test conditions, was assessed using Bland-Altman analysis. The error between data using the current one-size-fit-all correction and the gold-standard correction was compared with the error between the OAE-derived and the gold-standard correction. An individualised scaling factor based on the low frequency OAE information was derived for each participant. The error between data using the one-size-fit-all and gold-standard correction was compared with the error between the scaled correction and the gold-standard correction. Results: A strong agreement of 2.3 dB at 0.5 and 1 kHz was found between the gold-standard probe microphone RECD and the OAE based RECD in the low frequencies. The agreement in the mid and high frequencies is weaker with the agreement levels being 3.6 and 5.5 dB at 2 and 4 kHz, respectively. The mean error between the OAE based RECD and the average RECD was 1.8, 1.3, 1, and 3.4 dB at 0.5, 1, 2 and 4 kHz, respectively. Maximum error was 4.3, 3.7, 5.2, and 12.7 dB at 0.5, 1, 2 and 4 kHz, respectively. In 15 out of 20 cases, the scaled RECD based on the low frequency OAE probe data resulted in error either the same or less than what the average RECD offers. Conclusions: For adults, our data does not support the use of an OAE based individualised correction. The reduced agreement in the mid and higher frequencies compared to in the low frequencies is likely attributed to standing waves forming in the ear canal near the OAE microphone. The possibility of using the OAE based RECD cannot be ruled out for the infant population where standing waves have an effect at higher frequencies. The low frequency OAE information can be utilised to derive an individualised scaling factor to improve the accuracy of the average RECD that is employed in the one-size-fits-all nHL-to-eHL correction factor.